news2

Validating prediction models supplementing NEWS2 with blood parameters to predict 14-day ICU/death among COVID patients

Bridging the gap between theory and bedside practice through open-access, physiologically-weighted, evidence based, decision support tools for anesthesia and intensive care.

Real-time clinical vitals monitoring system that calculates NEWS2 scores, triggers tiered risk alerts, tracks temporal trends, and maps data using GDPR-compliant persistence

Three-tier hierarchical clinical AI for ESI triage prediction | OOF QWK 0.9987 | LightGBM + CatBoost + XGBoost ensemble with NEWS2, CCI, Modified Shock Index | Triagegeist Competition 2026

Interactive hospital capacity and clinical early-warning analytics in GNU Octave, driven live from a GitHub Copilot canvas. Runs on Synthea synthetic patient data with no MATLAB licence and no PHI.

Tracks patients in an Emergency Department from arrival to discharge. Auto-NEWS2, signed notes, AI clerking parser, AI discharge summaries. Live demo.

Using AI and UK clinical guidelines to predict sepsis onset for quicker detection and early management.

National Early Warning Score 2 implementation

A multi-agent clinical triage & escalation co-pilot that keeps the risk decision in deterministic code (NEWS2) and audits its own advice. Google ADK 2.0, MCP, Gemini, Cloud Run. Research prototype, not a medical device.

Clean-room healthcare signal analytics for edge, bedside, and remote monitoring. 63 Rust functions, 127 tests, Kalman filter, streaming API, criterion benchmarks. Zero ML dependencies.

Open pre-hospital alerting: validated clinical scores (NEWS2/BE-FAST/qSOFA), integrated ambulance→ED pre-alert, FHIR R4 export, hash-chained provenance (digest-only, privacy by design). Not a medical device.